Until now, in a scanner (image reading apparatus), or in a compound machine that installs a scanner, in the case to read manuscript, firstly, user sets a desirable scan resolution through an operation panel.
At this time, though it depends on a read manuscript image, if the resolution is too low, crush occurs in images of characters and segments that are obtained; on the contrary, if the resolution is higher than it is needed, scan time becomes long and capacity of read image data also becomes large (at this time, large capacity memory becomes necessary, and process time is also increased.). Therefore, it is inevitable for user to perform operation that spends time and labor so as to determine a proper resolution corresponding to manuscript after repeated trial and error, and is worst in convenience.
In view of such inconvenience, for example, in patent document 1, such technology is disclosed that once scans manuscript, determine a proper resolution through the image data, and scans again through the proper resolution. However, in the technology of the patent document 1, because it pre-scans through a highest resolution of the apparatus, the problem is still left that affects the scan time and the data capacity that are mentioned above.
Patent document 1: Japan patent publication of No. 2003-219150